Connect your agent. Inspect the execution record.

Submit runs, retrieve their status and inspect available artifacts through your existing MCP client. The tool list returned by the hosted service is the current capability reference.

A completed model run is not proof that a repository check ran. Inspect execution records and available verification results; shared hosted command verification is disabled by default.

Protocol version 2025-11-25 is the default; 2024-11-05, 2025-03-26, and 2025-06-18 are also supported.

Use an HTTP-capable MCP client. Create an AdaptOrch account and key, then configure the endpoint. Do not clone the private core for this path.

1 · Create a key

Create an account at /app/signup and create an API key at /app/api-keys. Use this value as ADAPTORCH_API_KEY throughout the direct hosted instructions. Keep actual keys out of committed files.

export ADAPTORCH_API_KEY="ado_..."

2 · Confirm authentication without running a model

This read-only tools/list request checks the endpoint and the supplied credentials. It does not submit a run, validate every client configuration, or prove that a code check is available.

: "${ADAPTORCH_API_KEY:?Set your AdaptOrch API key first}"
curl --fail --silent --show-error https://adaptorch.com/mcp \
  -H "Authorization: Bearer $ADAPTORCH_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Accept: application/json, text/event-stream" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}'

A valid key returns a result.tools array including adaptorch_run. Without a key the same call returns {"code":-32001,"message":"Missing bearer token"} with HTTP 401 — that response is itself proof the endpoint is reachable.

Set these only in your local environment; never in a shared transcript or committed file.

export ADAPTORCH_PROVIDER="openai"
export ADAPTORCH_PROVIDER_MODEL="YOUR_SUPPORTED_MODEL"
export ADAPTORCH_PROVIDER_KEY="YOUR_PROVIDER_KEY"

Start with the complete Claude Code HTTP example below. The other configuration blocks show client-specific authentication syntax only; add your provider headers using the authenticated dashboard guide before model execution.

Claude Code — one command

: "${ADAPTORCH_API_KEY:?Set your AdaptOrch API key}"
: "${ADAPTORCH_PROVIDER:?Set a supported provider}"
: "${ADAPTORCH_PROVIDER_MODEL:?Set the provider model ID}"
: "${ADAPTORCH_PROVIDER_KEY:?Set your provider key}"
claude mcp add --transport http adaptorch https://adaptorch.com/mcp \
  --header "Authorization: Bearer ${ADAPTORCH_API_KEY}" \
  --header "X-Provider: ${ADAPTORCH_PROVIDER}" \
  --header "X-Provider-Model: ${ADAPTORCH_PROVIDER_MODEL}" \
  --header "X-Provider-Key: ${ADAPTORCH_PROVIDER_KEY}"

Cursor — ~/.cursor/mcp.json

{
  "mcpServers": {
    "adaptorch": {
      "url": "https://adaptorch.com/mcp",
      "headers": { "Authorization": "Bearer ${ADAPTORCH_API_KEY}" }
    }
  }
}

Codex CLI — ~/.codex/config.toml (TOML, not JSON)

[mcp_servers.adaptorch]
url = "https://adaptorch.com/mcp"
bearer_token_env_var = "ADAPTORCH_API_KEY"
env_http_headers = { "X-Provider" = "ADAPTORCH_PROVIDER", "X-Provider-Model" = "ADAPTORCH_PROVIDER_MODEL", "X-Provider-Key" = "ADAPTORCH_PROVIDER_KEY" }

Gemini CLI — ~/.gemini/settings.json (key is httpUrl)

{
  "mcpServers": {
    "adaptorch": {
      "httpUrl": "https://adaptorch.com/mcp",
      "headers": { "Authorization": "Bearer ${ADAPTORCH_API_KEY}" }
    }
  }
}

VS Code — .vscode/mcp.json (top-level servers)

{
  "servers": {
    "adaptorch": {
      "type": "http",
      "url": "https://adaptorch.com/mcp",
      "headers": { "Authorization": "Bearer ${env:ADAPTORCH_API_KEY}" }
    }
  }
}

Windsurf — ~/.codeium/windsurf/mcp_config.json (key is serverUrl)

{
  "mcpServers": {
    "adaptorch": {
      "serverUrl": "https://adaptorch.com/mcp",
      "headers": { "Authorization": "Bearer ${ADAPTORCH_API_KEY}" }
    }
  }
}

Ask the connected client to call adaptorch_run once with the single-task payload below. This step can consume your plan allowance and your provider budget. Do not repeatedly submit it while polling.

{
  "payload": {"subtasks": [{"id": "ping", "description": "Reply exactly PONG", "estimated_tokens": 16}]},
  "synthesis_mode": "paper",
  "budget_policy": {"global_timeout_budget_seconds": 20}
}

Use the returned run_id with adaptorch_get_run and adaptorch_get_artifacts, or open /app/runs. Confirm the requested model and output. Then check whether any verification command actually ran; SUCCEEDED alone does not establish that.

401: check service authentication and provider-key requirements. 429: inspect quota/rate limits. Provider billing failures and unsupported verification environments are setup blockers, not successful activation.

Run records · Plan allowances and BYOK · Data flow

The local Python wrapper delegates to the AdaptOrch engine under proprietary licence terms. Downloading grants no broader licence; review the terms and the licence included with your distribution. This maintainer path requires the engine to be available and is not the hosted customer quickstart.

pip install adaptorch-mcp

export ADAPTORCH_CONTROL_PLANE_TOKEN="$ADAPTORCH_API_KEY"
export ADAPTORCH_MCP_PROVIDER="$ADAPTORCH_PROVIDER"
export ADAPTORCH_MCP_PROVIDER_MODEL="$ADAPTORCH_PROVIDER_MODEL"
export ADAPTORCH_MCP_PROVIDER_API_KEY="$ADAPTORCH_PROVIDER_KEY"
adaptorch-mcp --transport stdio --base-url https://adaptorch.com

For the local bridge, set ADAPTORCH_CONTROL_PLANE_TOKEN from your AdaptOrch API key. ADAPTORCH_MCP_PROVIDER, ADAPTORCH_MCP_PROVIDER_MODEL and ADAPTORCH_MCP_PROVIDER_API_KEY configure provider credentials in that process. Do not send secrets as tool arguments.

The hosted tools/list response is authoritative for the available tools. Results and errors follow the returned schemas. A tool-list response is not a model run or a verification result.

{ "jsonrpc": "2.0", "id": 1, "method": "tools/call",
  "params": { "name": "adaptorch_run",
    "arguments": { "prompt": "${TASK_PROMPT}", "synthesis_mode": "robust",
                   "wait_for_terminal": true, "timeout_seconds": 120.0 } } }

{ "jsonrpc": "2.0", "id": 2, "method": "tools/call",
  "params": { "name": "adaptorch_get_run", "arguments": { "run_id": "${RUN_ID}" } } }

{ "jsonrpc": "2.0", "id": 3, "method": "tools/call",
  "params": { "name": "adaptorch_get_artifacts", "arguments": { "run_id": "${RUN_ID}" } } }

{ "jsonrpc": "2.0", "id": 4, "method": "tools/call",
  "params": { "name": "adaptorch_list_runs", "arguments": { "limit": 20, "status": "SUCCEEDED" } } }

{ "jsonrpc": "2.0", "id": 5, "method": "tools/call",
  "params": { "name": "adaptorch_get_traces", "arguments": { "run_id": "${RUN_ID}" } } }

{ "jsonrpc": "2.0", "id": 6, "method": "tools/call",
  "params": { "name": "adaptorch_cancel_run", "arguments": { "run_id": "${RUN_ID}" } } }

{ "jsonrpc": "2.0", "id": 7, "method": "tools/call",
  "params": { "name": "adaptorch_route_topology",
    "arguments": { "subtasks": [ { "id": "t1" }, { "id": "t2" } ],
                   "dependencies": [ { "from": "t1", "to": "t2" } ] } } }

{ "jsonrpc": "2.0", "id": 8, "method": "tools/call",
  "params": { "name": "adaptorch_server_metrics", "arguments": {} } }

{ "jsonrpc": "2.0", "id": 9, "method": "tools/call",
  "params": { "name": "adaptorch_capabilities", "arguments": {} } }

{ "jsonrpc": "2.0", "id": 10, "method": "tools/call",
  "params": { "name": "adaptorch_plan_catalog", "arguments": {} } }

stdio relies on a local trust model plus a control-plane token. HTTP+SSE adds bearer auth, origin allow-listing, protocol-version checks, and a payload size cap. All tokens are injected through environment variables and are never hard-coded.

Three common flows cover most agent integrations.

1 · Run and collect

Call adaptorch_run with wait_for_terminal, then read the result with adaptorch_get_run, adaptorch_get_artifacts, and adaptorch_get_traces.

{ "name": "adaptorch_run",
  "arguments": { "prompt": "${TASK_PROMPT}", "wait_for_terminal": true } }

{ "name": "adaptorch_get_artifacts", "arguments": { "run_id": "${RUN_ID}" } }

2 · Pre-route a DAG

adaptorch_route_topology computes a recommended topology locally — it does not call the control-plane.

{ "name": "adaptorch_route_topology",
  "arguments": { "subtasks": [ { "id": "t1" }, { "id": "t2" } ],
                 "dependencies": [ { "from": "t1", "to": "t2" } ] } }

3 · Operate

List recent runs with adaptorch_list_runs and stop an in-flight run with adaptorch_cancel_run. The adaptorch_run_prompt and adaptorch_get_run_prompt prompts wrap these calls for agents.

• Launcher not found: use the adaptorch-mcp console script or scripts/run_adaptorch_mcp.sh.
• Token missing: set ADAPTORCH_CONTROL_PLANE_TOKEN; a remote --base-url requires it.
• Framing mismatch: match --stdio-framing (newline default vs content-length).
• HTTP 401 / 403: check the bearer token and the Origin allow-list.
• HTTP 413: the request body exceeds the payload size cap.
• verification_commands rejected: disabled unless ADAPTORCH_MCP_ALLOW_VERIFICATION_COMMANDS is set.
• Smoke test: GET /mcp/health and GET /mcp/sse?once=true.

• Adaptorch-MCP (GitHub)
• AdaptOrch MCP server (GitHub)
• Trust Center
• ingeng2004@gmail.com
• Open app